Michael D'Angelo
2b8a96907c
chore(release): shard verification tests and report release progress ( #1798 )
...
Run release test shards and MCP verification in parallel, then gate publication on every verification result. Maintain a release PR progress comment across CI, approvals, and publication, including recovery through follow-up fix commits.
Give parameterized tests unique JUnit names so baseline, parallel, and sharded inventories can be compared.
2026-10-21 12:16:11 +08:00
mldangelo-oai
59bb435f11
refactor: share workflow GitHub API calls ( #1466 )
2026-10-09 01:23:58 -04:00
mldangelo-oai and mldangelo-oai
2379b18126
refactor(tooling): remove duplicate benchmark checks and line collection ( #1359 )
...
refactor(tooling): reuse source discovery and benchmark validation
Co-authored-by: mldangelo-oai <269034524+mldangelo-oai@users.noreply.github.com >
2026-10-07 10:24:31 -04:00
mldangelo-oai
4cf2b14ffe
fix(scripts): run entrypoints through symlinked paths ( #1256 )
...
* fix(scripts): run entrypoints through symlinked paths
* test(scripts): pass preload modules as file URLs
* test: intercept fetch in script entrypoint fixtures
2026-10-05 15:32:00 -04:00
mldangelo-oai and Codex
b453780722
refactor(ci): consolidate build and tool setup ( #1204 )
...
CI workflows repeat package-manager, Node and build setup. Move the repeated steps into a local composite action and simplify equivalent build and package checks.
Co-authored-by: Codex <noreply@openai.com >
2026-10-03 15:06:17 -07:00
mldangelo-oai and mldangelo-oai
619d389c7f
feat(examples): add Invoice Desk fixture and OpenAI PR scans ( #1144 )
...
* feat(examples): add Invoice Desk fixture and OpenAI PR scans
* docs(examples): add generated Invoice Desk threat model
* ci(examples): diagnose OpenAI authentication after scan failures
* ci(examples): require protected approval for inference
* ci(examples): queue Invoice Desk scans for PRs to main
* ci(examples): persist CLI downloads and measure Invoice Desk scans
* fix(ci): preserve scan dispatches and complete source input
* fix(ci): cover automated PRs and retain incomplete scan metrics
* fix(ci): retain fallback for undispatched pull request heads
* fix(ci): remove sample metadata before inference
* fix(examples): deduplicate fallback scans and clarify setup
* fix(ci): isolate scan concurrency by revision
* ci(example): pass required workflow checks
* fix(example): improve text contrast
* test: cover sample scan workflow outputs
---------
Co-authored-by: mldangelo-oai <269034524+mldangelo-oai@users.noreply.github.com >
2026-10-01 15:29:56 -07:00
Kyle Brown
f1413cfbf8
refactor(ci): port source and test report checks to TypeScript ( #768 )
...
* refactor(ci): port source and test report checks to JavaScript
* refactor(ci): use TypeScript for migrated checks
* fix(ci): compile TypeScript checks before execution
* fix(test): give parameterized CI cases unique names
* fix(test): give release test cases unique names
* fix(test): give boolean matching cases unique report names
2026-09-08 21:16:31 +00:00
faizan-oai and faizan-oai
6f9257814d
fix(ci): preserve LF Python source checkouts ( #726 )
...
Co-authored-by: faizan-oai <269039902+faizan-oai@users.noreply.github.com >
2026-08-28 23:17:37 -07:00
soyeon-oai
047414649c
ci(plugin): enforce source import compatibility ( #722 )
...
* ci(plugin): check source import compatibility
* fix(plugin): close source compatibility gaps
* fix(plugin): parse markdown blocks accurately
* fix(plugin): cover markdown boundary cases
* fix(ci): keep source compatibility checks OSS-only
2026-08-28 15:36:12 -07:00